From: John Harrison <John.C.Harrison@Intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 02/34] drm/i915/execlists: Suppress preempting self
Date: Tue, 22 Jan 2019 14:18:46 -0800 [thread overview]
Message-ID: <3c659235-825f-bb6c-478b-6392d4541d12@Intel.com> (raw)
In-Reply-To: <20190121222117.23305-3-chris@chris-wilson.co.uk>
On 1/21/2019 14:20, Chris Wilson wrote:
> In order to avoid preempting ourselves, we currently refuse to schedule
> the tasklet if we reschedule an inflight context. However, this glosses
> over a few issues such as what happens after a CS completion event and
> we then preempt the newly executing context with itself, or if something
> else causes a tasklet_schedule triggering the same evaluation to
> preempt the active context with itself.
>
> To avoid the extra complications, after deciding that we have
> potentially queued a request with higher priority than the currently
> executing request, inspect the head of the queue to see if it is indeed
> higher priority from another context.
>
> References: a2bf92e8cc16 ("drm/i915/execlists: Avoid kicking priority on the current context")
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Can you explain what was wrong with the previous version of this patch
(drm/i915/execlists: Store the highest priority context)? It seemed simpler.
John.
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2019-01-22 22:18 UTC|newest]
Thread overview: 89+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-21 22:20 HWSP for HW semaphores Chris Wilson
2019-01-21 22:20 ` [PATCH 01/34] drm/i915/execlists: Mark up priority boost on preemption Chris Wilson
2019-01-21 22:20 ` [PATCH 02/34] drm/i915/execlists: Suppress preempting self Chris Wilson
2019-01-22 22:18 ` John Harrison [this message]
2019-01-22 22:38 ` Chris Wilson
2019-01-21 22:20 ` [PATCH 03/34] drm/i915: Show all active engines on hangcheck Chris Wilson
2019-01-22 12:33 ` Mika Kuoppala
2019-01-22 12:42 ` Chris Wilson
2019-01-21 22:20 ` [PATCH 04/34] drm/i915/selftests: Refactor common live_test framework Chris Wilson
2019-01-22 12:37 ` Matthew Auld
2019-01-21 22:20 ` [PATCH 05/34] drm/i915/selftests: Track evict objects explicitly Chris Wilson
2019-01-22 11:53 ` Matthew Auld
2019-01-21 22:20 ` [PATCH 06/34] drm/i915/selftests: Create a clean GGTT for vma/gtt selftesting Chris Wilson
2019-01-22 12:07 ` Matthew Auld
2019-01-21 22:20 ` [PATCH 07/34] drm/i915: Refactor out intel_context_init() Chris Wilson
2019-01-22 12:32 ` Matthew Auld
2019-01-22 12:39 ` Mika Kuoppala
2019-01-22 12:48 ` Chris Wilson
2019-01-21 22:20 ` [PATCH 08/34] drm/i915: Make all GPU resets atomic Chris Wilson
2019-01-22 22:19 ` John Harrison
2019-01-22 22:27 ` Chris Wilson
2019-01-23 8:52 ` Mika Kuoppala
2019-01-21 22:20 ` [PATCH 09/34] drm/i915/guc: Disable global reset Chris Wilson
2019-01-22 22:23 ` John Harrison
2019-01-21 22:20 ` [PATCH 10/34] drm/i915: Remove GPU reset dependence on struct_mutex Chris Wilson
2019-01-24 12:06 ` Mika Kuoppala
2019-01-24 12:50 ` Chris Wilson
2019-01-24 13:12 ` Chris Wilson
2019-01-24 14:10 ` Chris Wilson
2019-01-21 22:20 ` [PATCH 11/34] drm/i915/selftests: Trim struct_mutex duration for set-wedged selftest Chris Wilson
2019-01-21 22:20 ` [PATCH 12/34] drm/i915: Issue engine resets onto idle engines Chris Wilson
2019-01-23 1:18 ` John Harrison
2019-01-23 1:31 ` Chris Wilson
2019-01-21 22:20 ` [PATCH 13/34] drm/i915: Stop tracking MRU activity on VMA Chris Wilson
2019-01-21 22:20 ` [PATCH 14/34] drm/i915: Pull VM lists under the VM mutex Chris Wilson
2019-01-22 9:09 ` Tvrtko Ursulin
2019-01-21 22:20 ` [PATCH 15/34] drm/i915: Move vma lookup to its own lock Chris Wilson
2019-01-21 22:20 ` [PATCH 16/34] drm/i915: Always allocate an object/vma for the HWSP Chris Wilson
2019-01-21 22:21 ` [PATCH 17/34] drm/i915: Move list of timelines under its own lock Chris Wilson
2019-01-21 22:21 ` [PATCH 18/34] drm/i915/selftests: Use common mock_engine::advance Chris Wilson
2019-01-22 9:33 ` Tvrtko Ursulin
2019-01-21 22:21 ` [PATCH 19/34] drm/i915: Tidy common test_bit probing of i915_request->fence.flags Chris Wilson
2019-01-22 9:35 ` Tvrtko Ursulin
2019-01-21 22:21 ` [PATCH 20/34] drm/i915: Introduce concept of per-timeline (context) HWSP Chris Wilson
2019-01-23 1:35 ` John Harrison
2019-01-21 22:21 ` [PATCH 21/34] drm/i915: Enlarge vma->pin_count Chris Wilson
2019-01-21 22:21 ` [PATCH 22/34] drm/i915: Allocate a status page for each timeline Chris Wilson
2019-01-21 22:21 ` [PATCH 23/34] drm/i915: Share per-timeline HWSP using a slab suballocator Chris Wilson
2019-01-22 10:47 ` Tvrtko Ursulin
2019-01-22 11:12 ` Chris Wilson
2019-01-22 11:33 ` Tvrtko Ursulin
2019-01-21 22:21 ` [PATCH 24/34] drm/i915: Track the context's seqno in its own timeline HWSP Chris Wilson
2019-01-22 12:24 ` Tvrtko Ursulin
2019-01-21 22:21 ` [PATCH 25/34] drm/i915: Track active timelines Chris Wilson
2019-01-22 14:56 ` Tvrtko Ursulin
2019-01-22 15:17 ` Chris Wilson
2019-01-23 22:32 ` John Harrison
2019-01-23 23:08 ` Chris Wilson
2019-01-21 22:21 ` [PATCH 26/34] drm/i915: Identify active requests Chris Wilson
2019-01-22 15:34 ` Tvrtko Ursulin
2019-01-22 15:45 ` Chris Wilson
2019-01-21 22:21 ` [PATCH 27/34] drm/i915: Remove the intel_engine_notify tracepoint Chris Wilson
2019-01-22 15:50 ` Tvrtko Ursulin
2019-01-23 12:54 ` Chris Wilson
2019-01-23 13:18 ` Tvrtko Ursulin
2019-01-23 13:24 ` Chris Wilson
2019-01-21 22:21 ` [PATCH 28/34] drm/i915: Replace global breadcrumbs with per-context interrupt tracking Chris Wilson
2019-01-23 9:21 ` Tvrtko Ursulin
2019-01-23 10:01 ` Chris Wilson
2019-01-23 16:28 ` Tvrtko Ursulin
2019-01-23 11:41 ` [PATCH] " Chris Wilson
2019-01-21 22:21 ` [PATCH 29/34] drm/i915: Drop fake breadcrumb irq Chris Wilson
2019-01-24 17:55 ` Tvrtko Ursulin
2019-01-24 18:18 ` Chris Wilson
2019-01-21 22:21 ` [PATCH 30/34] drm/i915: Keep timeline HWSP allocated until the system is idle Chris Wilson
2019-01-21 22:37 ` Chris Wilson
2019-01-21 22:48 ` Chris Wilson
2019-01-21 22:21 ` [PATCH 31/34] drm/i915/execlists: Refactor out can_merge_rq() Chris Wilson
2019-01-21 22:21 ` [PATCH 32/34] drm/i915: Use HW semaphores for inter-engine synchronisation on gen8+ Chris Wilson
2019-01-21 22:21 ` [PATCH 33/34] drm/i915: Prioritise non-busywait semaphore workloads Chris Wilson
2019-01-23 0:33 ` Chris Wilson
2019-01-21 22:21 ` [PATCH 34/34] drm/i915: Replace global_seqno with a hangcheck heartbeat seqno Chris Wilson
2019-01-22 0:09 ` ✗ Fi.CI.CHECKPATCH: warning for series starting with [01/34] drm/i915/execlists: Mark up priority boost on preemption Patchwork
2019-01-22 0:22 ` ✗ Fi.CI.SPARSE: " Patchwork
2019-01-22 0:30 ` ✓ Fi.CI.BAT: success " Patchwork
2019-01-22 1:35 ` ✗ Fi.CI.IGT: failure " Patchwork
2019-01-23 12:00 ` ✗ Fi.CI.CHECKPATCH: warning for series starting with [01/34] drm/i915/execlists: Mark up priority boost on preemption (rev2) Patchwork
2019-01-23 12:11 ` ✗ Fi.CI.SPARSE: " Patchwork
2019-01-23 12:48 ` ✗ Fi.CI.BAT: failure " Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=3c659235-825f-bb6c-478b-6392d4541d12@Intel.com \
--to=john.c.harrison@intel.com \
--cc=chris@chris-wilson.co.uk \
--cc=intel-gfx@lists.freedesktop.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.